Magnesium L-Threonate Benefits: What the Research Actually Shows

TL;DR

Magnesium L-threonate is the form developed to get more magnesium into the brain. The evidence for that specific claim comes from animal studies, most of them from the lab that created the compound. Three human trials have tested it for cognition, and the cognitive results hold up better than the sleep results. Every one of those human trials was funded by a company with a commercial interest in the ingredient. Human evidence here is early, and the honest read is cautious optimism rather than confidence.

What magnesium L-threonate is

Magnesium L-threonate pairs magnesium with L-threonic acid, a metabolite of vitamin C. It was created in 2010 by a research group at MIT and Tsinghua, and the branded version used in most published trials is called Magtein.

The compound is mostly threonate by weight. The European Food Safety Authority assessed it as a novel food in 2024 and noted that a 3,000 mg daily dose of the compound delivers around 250 mg of magnesium. That puts the elemental magnesium content near 8 percent, which is low next to other supplement forms (EFSA 2024). The same panel concluded that magnesium is bioavailable from this source, and that the intake levels proposed for supplements are safe.

Read a threonate label with that ratio in mind. A gram of the compound is a few dozen milligrams of actual magnesium.

The brain magnesium claim, and where it comes from

The founding study appeared in Neuron in 2010. Researchers raised brain magnesium in rats using the new compound and reported better learning, working memory, and short and long term memory. Aged rats improved on pattern completion. The treated animals had a higher density of synaptophysin and synaptobrevin positive puncta in the hippocampus, and that density correlated with memory improvement (Slutsky 2010).

That is rat data. The last author, Guosong Liu, developed the compound and went on to found the company that commercialized it. The finding is real and it has been cited widely. It is also not a human result, and it gets repeated as though it were one.

A 2016 follow-up from the same group worked out the mechanism. Threonate turns out to occur naturally in cerebrospinal fluid, and oral dosing raised CSF threonate levels. In cultured hippocampal neurons, threonate itself drove magnesium into the cell through glucose transporters, upregulated NR2B-containing NMDA receptors, and raised functional synapse density. Other magnesium anions did not do this (Sun 2016). The work spans rat tissue, cell culture, and human stem cell derived neurons. None of it involves people taking a supplement.

An independent group tested the delivery question directly in mice. They gave magnesium L-threonate or magnesium sulfate by mouth for four weeks and measured magnesium in serum and cerebrospinal fluid. CSF magnesium rose in the threonate group. It did not rise with magnesium sulfate (Shen 2019).

Three studies, three species, zero humans. No published trial has measured brain or cerebrospinal fluid magnesium in people taking this supplement.

What the human trials found on cognition

The first human trial ran in 2016. Forty-four older adults aged 50 to 70 with cognitive impairment took the compound or placebo for 12 weeks. Overall cognitive ability improved against placebo with a large effect size (p = 0.003, Cohen's d = 0.91), and the group's impaired executive function came close to normal range (Liu 2016). Guosong Liu was first author again.

A 2026 trial in Frontiers in Nutrition is the largest and best-instrumented so far. One hundred adults aged 18 to 45 with self-reported poor sleep took 2 g daily of the branded form or placebo for six weeks. Overall cognition improved on the NIH Cognitive Toolbox (p = 0.043), with the effect concentrated in working and episodic memory. Reaction time improved (p = 0.031). Raven's Progressive Matrices showed nothing (p = 0.953) (Lopresti 2026). The study was funded by Threotech, which supplied the product and helped design it.

A third trial in 109 healthy Chinese adults reported gains across all five subcategories of a standard memory test after 30 days (Zhang 2022). The product tested there combined magnesium L-threonate with phosphatidylserine plus vitamins C and D. Phosphatidylserine has its own cognitive research, so this trial cannot tell you what the threonate contributed. The funding came from a magnesium L-threonate manufacturer.

What the human trials found on sleep

This is the part that gets oversold.

The 2016 trial set out to measure sleep alongside cognition. The authors reported that placebo effects on sleep and anxiety were strong enough that the compound's effect on either one could not be determined (Liu 2016). That's a null result on the sleep question, stated by the researchers themselves.

A 2024 trial put 80 adults aged 35 to 55 with self-reported sleep problems on 1 g daily for 21 days. Deep sleep score, REM sleep score, and light sleep time improved against placebo on an Oura ring, along with several subjective measures. The authors describe the pattern as the treatment group holding steady while the placebo group declined (Hausenblas 2024). Three of the six authors work for AIDP, which supplies the ingredient, and the paper carries a published corrigendum (erratum 2025).

The 2026 trial ran longer, in more people, with the same category of wearable. On subjective measures, sleep-related impairment improved (p = 0.043). Sleep disturbances did not (p = 0.316). Restorative sleep did not (p = 0.439). On the objective ring data, there were no group differences in any sleep outcome at all. Resting heart rate fell and heart rate variability rose, which the authors read as a stress signal rather than a sleep one (Lopresti 2026).

Two trials with wearable sleep tracking, pointing opposite directions on the objective measures. The larger and longer one found nothing.

Other human evidence

Outside cognition and sleep, the strongest human trial is in pain. Eighty-three patients with advanced cancer taking oral opioids received 1.5 to 2.0 g of the compound or placebo. The treatment group needed smaller opioid dose increases over 90 days, and reported less opioid-induced constipation (Wu 2023). That population is far from a general sleep audience, and it says nothing about sleep quality in healthy adults. It does show the compound doing something measurable in people.

Where the hype outruns the data

A claim you will see repeated everywhere is that most magnesium forms have poor bioavailability and threonate is the exception. That line appears in industry-funded papers on the ingredient. It does not survive contact with the wider literature. A review of magnesium absorption concluded that the type of magnesium salt matters less than dose and existing magnesium status (Schuchardt and Hahn 2017), and EFSA rated threonate as a bioavailable source without ranking it above the others (EFSA 2024).

The second overstatement is treating the rat synapse work as though it happened in people. Higher synaptic density in a rat hippocampus is a striking result. It is not a measured outcome in any human taking this supplement.

The third is the funding pattern. All three human cognition trials were paid for by companies selling the ingredient. Industry funding is normal in nutrition research and the results can still be sound. What's missing is the independent replication that would settle the question.

How it compares to glycinate

Glycinate has more human data behind it for the things most people buy magnesium for, and it delivers far more elemental magnesium per gram. Threonate has the more interesting mechanism and a thinner human file.

If your goal is sleep, the general magnesium evidence sits on firmer ground than the threonate-specific evidence. A trial in 46 older adults with insomnia found improvements in sleep time, sleep efficiency, and sleep onset latency on 500 mg of magnesium daily (Abbasi 2012). A meta-analysis of three trials in 151 older adults found sleep onset latency 17.4 minutes shorter than placebo, while rating the evidence low to very low quality (Mah and Pitre 2021). Neither involved threonate.

If your goal is memory or mental sharpness, threonate is the form with human cognition trials attached to it, thin and industry-funded as they are.

For the side by side on the gentler everyday forms, see magnesium glycinate vs citrate and magnesium glycinate benefits. For a walk through every form at once, see which magnesium is best for sleep and anxiety.

FAQ

Is L-threonate better than glycinate?

For sleep, no trial has compared them, and the general magnesium sleep evidence was gathered with other forms. For cognition, threonate is the only form with human trials aimed at that outcome, though all three were funded by ingredient suppliers (Liu 2016, Lopresti 2026). Glycinate gives you several times more elemental magnesium per gram.

Does it cross the blood brain barrier?

In mice, oral magnesium L-threonate raised cerebrospinal fluid magnesium while magnesium sulfate did not (Shen 2019). In rats, the compound raised brain magnesium and synapse density (Slutsky 2010). Cell work suggests threonate itself carries magnesium into neurons through glucose transporters (Sun 2016). Nobody has measured this in a living person. Anyone telling you it's proven in humans is filling a gap with confidence.

What dose is used in studies?

Human trials have used 1 g to 2 g of the compound daily. The 2024 sleep trial used 1 g for 21 days (Hausenblas 2024). The 2026 cognition and sleep trial used 2 g for six weeks (Lopresti 2026). The 2023 cancer pain trial used 1.5 to 2.0 g by body weight (Wu 2023). EFSA reviewed intakes up to 3 g of the compound daily and found them safe (EFSA 2024). Doses well below 1 g have not been tested for these outcomes.

Is Magtein the same thing?

Magtein is a branded form of magnesium L-threonate, and it's the version used in most of the published human trials. The underlying compound is the same molecule whichever name is on the box. Trials run on the branded material are the ones with published results.

Is it worth taking?

If you want magnesium for sleep or general repletion, cheaper forms deliver more elemental magnesium and have more human data. If you're specifically interested in the cognitive angle and you accept that the human evidence is early and industry-funded, threonate is the form those trials used. Set your expectations to match the evidence, which is promising and unfinished.

Disclaimer

This article is for general information only. It is not medical advice and it does not replace guidance from a qualified healthcare professional. Magnesium supplements can interact with certain medications, including some antibiotics and diuretics. People with reduced kidney function should not take supplemental magnesium without medical supervision. Do not use this information to diagnose or treat any condition. Speak with your doctor before starting any supplement.